Python Weekly (Issue 689 March 6 2025)

Welcome to issue 689 of Python Weekly. I have a special announcement to make below.


Join Our Angel Syndicate: Invest in the Future of Tech


I run an angel syndicate investing in early-stage tech startups, backing visionary founders with capital and support.


For Investors:

  • No Cost to Join: Become part of the syndicate with no upfront costs.
  • Flexible Investing: Invest deal by deal with no obligation to commit beyond your chosen deals.
  • Low Minimums: Access curated opportunities without high entry barriers.
  • Accredited Investors Only: Participation is limited to accredited investors.

For Founders:

  • Building a venture-scale business? We’re eager to hear from ambitious tech founders solving big problems and creating innovative solutions.

Reply back to this email to connect or learn more!


Articles, Tutorials and Talks

This video explores the key differences between TypeScript and Python, focusing on TypeScript from the perspective of a Python developer. It covers topics like type systems, interfaces, callables, and the ecosystem, providing insights into how a Pythonista can approach TypeScript.

This article describes an informal "study" testing the idea that kebabs near Paris train stations are worse. Using data from OSMnx and the Google Places extension, it found no link between a kebab shop's distance from a station and its Google review rating.

The author automated podcast transcription using roboscribe, a Python tool that combines WhisperX for diarized transcription and a local Large Language Model (LLM) for cleaning up the transcript, significantly improving readability. By leveraging local AI models, the author maintains control and optimizes the transcription process on their own hardware, achieving high-quality results in a fraction of the time compared to manual editing.

This blog post explores implementing typeclasses and overloading in Knuckledragger, a Python-based proof assistant using Z3. It discusses managing abstraction layers between Python and Z3 expressions using SortDispatch and dataclasses.

The course starts with the fundamentals, covering what machine learning is, how it differs from traditional approaches, and where it’s used. It then dives deeper into the mechanics, exploring different models, algorithms, and training processes. Next, it introduces Generative AI, explaining how it creates new content, before wrapping up with the architecture of AI systems and how to design and deploy them effectively.

Python has a built-in help function for getting help... but what can do with help?

This article introduces a new, low-overhead method for memory profiling in PyPy by combining VMProf and the PyPy Garbage Collector. It samples memory allocations by integrating the sampling check into the bump pointer allocator check, reducing overhead and allowing adjustable precision.


Start learning AI in 2025

Keeping up with AI is hard – we get it!

That’s why over 1M professionals read Superhuman AI to stay ahead.

  • Get daily AI news, tools, and tutorials

  • Learn new AI skills you can use at work in 3 mins a day

  • Become 10X more productive


Interesting Projects, Tools, and Libraries

$100K or 100 Days: Trade-offs when Pre-Training with Academic Resources.

Manage GPU clusters for running AI models.

A cutting-edge speech restoration model designed to significantly enhance the quality of degraded voice recordings.

NotaGen is a symbolic music generation model leveraging Large Language Models (LLMs) through pre-training on 1.6M musical pieces, fine-tuning on classical compositions, and reinforcement learning using a novel CLaMP-DPO method.

A toolkit for training language models to work with PDF documents in the wild.

A lightweight data processing framework built on DuckDB and 3FS.

Open and Advanced Large-Scale Video Generative Models.

An Efficient, Scalable RL Training Framework for Reasoning & Search Engine Calling interleaved LLM based on veRL.

A text-to-speech (TTS) and Speech-to-Speech (STS) library built on Apple's MLX framework, providing efficient speech synthesis on Apple Silicon.

Prevent merging of malicious code in pull requests.


New Releases

This release includes the following announcements:

  • Improvements to Python shell integration

  • New setting for auto test discovery file pattern

  • Inline values on hover in Notebooks and Python files

  • Automatic quotation insertion when breaking long strings


Upcoming Events and Webinars

There will be following talks

  • McCache - The cluster aware local cache

  • Unlock proven strategies for building enterprise-scale RAG systems and learn from real-world success stories.

There will be following talks

  • The Duality of Python

  • Taking agents from demo to prod

There will be following talks

  • Symbolic Regression

  • Mind the Gap: Making Technical Concepts Click with Any Audience

There will be a workshop on Self-Hosting LLMs.

There will be a talk, Training Speech Recognition and Generation Models at Scale.


Our Other Newsletters

Programmer Weekly - A free weekly newsletter for programmers.

Founder Weekly - A free weekly newsletter for entrepreneurs featuring best curated content, must read articles, how to guides, tips and tricks, resources, events and more.